The GitHub Actions job "Java CI with Maven" on stormcrawler.git/main has 
succeeded.
Run started by GitHub user rzo1 (triggered by rzo1).

Head commit for run:
ead30fc90a77047342ed6114a996b4e890364403 / Abhinav <[email protected]>
Upgrade to Tika 4.0.0 (#2118)

* Upgrade to Tika 4.0.0

Closes #2063

Tika 4.0.0 requires several API and configuration changes:

- core: TikaConfig has been removed - JSoupParserBolt now uses
  DefaultDetector for mime type detection; Detector.detect takes a
  TikaInputStream and a ParseContext; the CONTENT_TYPE and
  CONTENT_LENGTH constants moved from Metadata to HttpHeaders.
- external/tika: ParserBolt loads configurations with TikaLoader
  (tika-serialization) since XML configurations are no longer
  supported, and the bundled tika-config.xml has been migrated to
  JSON. Config resources bundled in a jar are copied to a temporary
  file as TikaLoader reads configs from the filesystem only.
- external/tika: tika-parsers-standard-package is now a POM
  aggregating the individual parser modules and is consumed as such.
- the default value of parser.tika.config.file is now
  tika-config.json
- THIRD-PARTY.txt regenerated

* Address review feedback on the Tika 4 upgrade

- seed each parse with the ParseContext from the 'parse-context'
  section of the Tika configuration instead of ignoring it
- fail fast when the Tika configuration is missing or cannot be
  loaded: silently degrading to the default configuration would
  activate parsers the configuration excluded, e.g. OCR
- delete the temporary copy of the configuration right after
  TikaLoader has read it instead of keeping it for the lifetime
  of the worker
- bind EmptyParser in the ParseContext unless parser.extract.embedded
  is set: the AutoDetectParser sets itself as the embedded parser
  otherwise, so embedded documents were parsed even though the
  configuration said not to
- use org.apache.tika.metadata.HttpHeaders constants for Tika
  metadata calls
- fix the tika config key in the archetype crawler-conf.yaml files
  (parser.tika.config.file, not parse.tika.config.file) and point
  the README at a working Tika 4 documentation URL
- new TikaConfigTest covering the exclusions from the bundled
  configuration, the parse-context seeding and the fail-fast
  behaviour, plus a regression test for the embedded documents

* Tika 4 docs: document parser.extract.embedded and parse.* key renames

Report URL: https://github.com/apache/stormcrawler/actions/runs/34451792509

With regards,
GitHub Actions via GitBox

Reply via email to